The focus of this chapter has been on estimating changes in condition-related costs.
As we have shown, changes in condition-related costs are estimated based on esti- mated changes in health outcomes associated with the underlying condition and the associated use of a variety of health-care resources needed for the treatment of the underlying condition. Condition-related changes in these health outcomes and the use of health-care resources that are expected to occur during the budget-impact analysis time horizon can be calculated in addition to their use in estimating the changes in treatment-related costs. For example changes in the number of deaths or symptom days and in the number of physician visits or hospital days for the reimbursement- eligible population can be calculated for each year of the budget- impact analysis. They can be presented to decision makers to help them with assess- ing their progress toward meeting health targets and/or planning their resource needs (see Chap. 11).
Exercises
Exercise 6.1 Choose 10 conditions for which a new drug has come to market or is entering the market. Assuming that condition-related outcomes and costs are of interest to the audience for the model, determine whether these should be included
100
in the budget-impact analysis (i.e., whether they are expected to change on introduc- tion of the new for each drug). If so, what condition-related outcomes and/or costs should be considered for each drug?
Exercise 6.2 Consider a new drug coming to market to treat asthma. The pri- mary endpoint within the clinical trials is change in forced expiratory volume in the first second (FEV1). What condition-related costs might be considered in a budget- impact analysis, and how would you estimate these costs?
Exercise 6.3 A new drug is entering the market to treat a rare condition. There are few treatments in the market to treat this condition, and the endpoints within the clinical trials are based on clinical and laboratory evaluations (they are not clinical events that require resource utilization). How would you estimate condition-related outcomes and/or costs? Would you consider these types of costs and outcomes within a budget-impact analysis? Why or why not?
Exercise 6.4 A drug is being studied to replace surgery. How might a budget- impact analysis be set up, and how would condition-related outcomes and costs be considered?
Exercise 6.5 Another HMG-CoA reductase inhibitor (i.e., a statin) is expected to come to market to treat patients with high cholesterol. This statin is the eighth statin to market, but it has a cleaner side effect profile than existing drugs. How might a budget-impact analysis be set up to assess the impact of this new treat- ment? Would condition-related outcomes and/or costs be considered? Why or why not?
Exercise 6.6 A new treatment is coming to market. The manufacturer believes the treatment will reduce the length of stay in the hospital, but there is no good evi- dence of this compared with the other competing treatments. How might a study be designed to collect these condition-related outcomes and/or costs?
Exercise 6.7 Choose a condition for which a new treatment has come to mar- ket. Determine the condition-related costs that the new treatment might affect and collect the evidence for a budget-impact analysis.
Exercise 6.8 For the treatment in Exercise 6.7, design an Excel worksheet to present the condition-related outcomes and costs.
Exercise 6.9 Identify a condition for which a micro-costing approach might be necessary for presenting the impact on condition-related costs. Create an Excel worksheet to present this micro-costing.
Exercise 6.10 A budget-impact analysis is being built to examine annual bud- gets up to 5 years for a chronic condition in which the impact on resource use and costs is not expected until after the analysis time horizon. Explain why condition- related outcomes and costs might/might not be important to consider.
References
Colquitt JL, Jones J, Tan SC, Takeda A, Clegg AJ, Price A. Ranibizumab and pegaptanib for the treatment of age-related macular degeneration: a systematic review and economic evaluation.
Health Technol Assess. 2008;12(16):iii–v. ix–201
S. Wolowacz et al.
Freemantle N, Cooper C, Diez-Perez A, Gitlin M, Radcliffe H, Shepherd S, et al. Results of indi- rect and mixed treatment comparison of fracture efficacy for osteoporosis treatments: a meta- analysis. Osteoporos Int. 2013;24(1):209–17. doi:10.1007/s00198-012-2068-9.
Maggioni AP, Anand I, Gottleib SO, Latini R, Tognoni G. Cohn JN; Val_HeFT Investigators (Valsartan Heart Failure Trial). Effects of valsartan on morbidity and mortality in patients with heart failure not receiving angiotensin-converting enzyme inhibitors. J Am Coll Cardiol.
2002;40(8):1414–21.
Mahoney EM, Wang K, Arnold SV, Proskorovsky I, Wiviott S, Antman E, et al. Cost-effectiveness of prasugrel versus clopidogrel in patients with acute coronary syndromes and planned percutane- ous coronary intervention: results from the trial to assess improvement in therapeutic outcomes by optimizing platelet inhibition with prasugrel-thrombolysis in myocardial infarction TRITON- TIMI 38. Circulation. 2010;121(1):71–9. doi:10.1161/CIRCULATIONAHA.109.900704.
Mauskopf J, Chirila C, Birt J, Boye KS, Bowman L. Drug reimbursement recommendations by the National Institute for Health and Clinical Excellence: have they impacted the National Health Service budget? Health Policy. 2013;110(1):49–59. doi:10.1016/j.healthpol.2013.01.017.
Nicholson KG, Aoki FY, Osterhaus AD, Trottier S, Carewicz O, Mercier CH, et al. Efficacy and safety of oseltamivir in treatment of acute influenza: a randomized controlled trial.
Neuraminidase Inhibitor Flu Treatment Investigator Group. Lancet. 2000;355(9218):
1845–50.
Roskell NS, Setyawan J, Zimovetz EA, Hodgkins P. Systematic evidence synthesis of treatments for ADHD in children and adolescents: indirect treatment comparisons of lisdexamfetamine with methylphenidate and atomoxetine. Curr Med Res Opin. 2014;30(8):1673–85.
doi:10.1185/03007995.2014.904772.
Roskell NS, Zimovetz EA, Rycroft CE, Eckert BJ, Tyas DA. Annualized relapse rate of first-line treatments for multiple sclerosis: a meta-analysis, including indirect comparisons versus fingo- limod. Curr Med Res Opin. 2012;28(5):767–80. doi:10.1185/03007995.2012.681637.
Sullivan SD, Mauskopf JA, Augustovski F, Caro JJ, Lee KM, Minchin M, et al. Budget impact analysis—principles of good practice: report of the ISPOR 2012 Budget impact analysis good practice II task force. Value Health. 2014;17(1):5–14. doi:10.1016/j.jval.2013.08.2291.
103
© Springer International Publishing AG 2017
J. Mauskopf et al., Budget-Impact Analysis of Health Care Interventions, DOI 10.1007/978-3-319-50482-7_7
Chapter 7
The Computing Framework and Calculations
Anita Brogan, Stephanie Earnshaw, and Josephine Mauskopf
Abstract The process of developing the computing framework for a budget-impact analysis includes making decisions about the computer model structure while accounting for available input data, necessary assumptions, and other model design requirements. The computing framework can be developed once the analytic framework for the budget-impact analysis has been determined, a detailed flow diagram has been created, and the methods for estimating input parameter values including population size, treatment mix, treatment-related costs, and condition- related costs have been determined. In general, a static cost-calculator model struc- ture should be used whenever such a model can credibly capture the impact of the new drug on the decision maker’s budget. However, there are circumstances when accurately estimating budget impact requires the use of more complex calcula- tions, such as the use of formal decision-analytic modeling techniques. These cir- cumstances generally occur when a dynamic accounting of the treated population is necessary.
Keywords Computing framework • Budget-impact analysis • Cost calculator
• Decision-analytic model • Static • Dynamic
Chapter Goal
To show, with examples, that once the researcher has developed the analytic framework and determined the availability of data for estimating population size and relevant descriptors, treatment mix and costs, and condition- related outcomes and costs, decisions can be made about the computing framework.
A. Brogan (*) • S. Earnshaw • J. Mauskopf RTI Health Solutions,
RTI International, Research Triangle Park, Durham, NC, USA e-mail: abrogan@rti.org
The process of developing the computing framework for a budget-impact analy- sis for a new drug1 includes making decisions about the model structure while accounting for available input data, necessary assumptions, and other model design requirements. The computing framework can be developed once the analytic framework for the budget-impact analysis has been determined, a detailed flow dia- gram has been created, and the methods for estimating input parameter values including population size, treatment mix, treatment-related costs, and condition- related costs have been determined.
Since the primary purpose of a budget-impact analysis is to help health care budget holders plan their budgets over their typical planning time horizons, it is critical that the computing framework be kept as simple as possible. Simplicity helps ensure that the analysis is transparent and can be readily adapted to the envi- ronment specific to each budget holder. The National Institute for Health and Care Excellence (NICE) costing templates are good examples of budget-impact analyses that use simple cost-calculator computing frameworks. However, more complex computing frameworks that include disease progression models are sometimes pre- ferred when a combined model is desired to assess both the budget impact and the cost- effectiveness of the new intervention or when a cost-calculator computing framework cannot credibly capture the changes in population size, treatment pat- terns, or condition outcomes over the appropriate time horizon.
Developing the computing framework may feel like putting together a puzzle.
The pieces of the puzzle have been identified, and now it is time to put them together to form one cohesive picture. Through this process, the overall size of the puzzle will be determined, and it may become apparent that a few additional pieces are needed or that a few extra pieces can be left out. When the puzzle is complete, the final version should provide a clear picture of the impact of the new drug on annual budgets as well as on annual resource use and/or health outcomes.
When making decisions about the computing framework for the budget-impact analysis, the target user should always be the driving factor. Decision makers can only use information from a budget-impact analysis if they find the analysis and results to be credible. This concept is critical to keep in mind because if any aspect of the analysis does not align with the decision maker’s expectations, the analysis may be rejected without further consideration. Therefore, all decisions should be made while carefully considering the decision maker’s perspective. Ideally, the planned computing framework and all assumptions and input parameter values can be shared with appropriate stakeholders early in the process to obtain feedback and assess credibility (see Chap. 9 on face validity).
The computing framework should be developed to allow the analysis to capture the relevant population(s), costs, and resource use and/or health outcomes while remaining as simple as possible. In this chapter, we provide an overview of budget- impact calculations, discuss static versus dynamic approaches to these calculations, and provide simple examples.
1 In this chapter, we make the simplifying assumption that the budget-impact analysis is based on the introduction of a new drug to the current mix of drugs for the treatment of a condition. Changes in our recommended approaches to estimate the budget impacts of other types of health care inter- ventions (i.e., vaccines, diagnostics, surgery, and devices) are discussed in Chap. 13.
105